centos下的MYSQL问题

hackroad

服务器日发帖量5000左右 PV 11W左右
服务器为centos
PHP v5.1.6
lighttpd/1.4.18
MYSQL 5.0.45

MYSQL 数据库 1.38 GB
独立服务器

服务器负载了一个官方论坛 使用discuz程序
使用xshow首页四格插件

设置xshow 缓存时间为60秒,如果设置为0 每次刷新首页会造成MYSQL CPU占用率极高
无奈设置了60秒,如果设置过高那就失去了这个首页四格的意义.
服务器内存加到了4G 效果稍微改善,但仅仅这个论坛不可能造成服务器这么卡.
现在设置60秒后 仍然间接性发帖 打开首页 回帖 卡住. 卡肯定是卡在MYSQL上了.

目前深夜服务器负载图

1. # Example MySQL config file for large systems.
   
2. #
   
3. # This is for a large system with memory = 512M where the system runs mainly
   
4. # MySQL.
   
5. #
   
6. # You can copy this file to
   
7. # /etc/my.cnf to set global options,
   
8. # mysql-data-dir/my.cnf to set server-specific options (in this
   
9. # installation this directory is /var/lib/mysql) or
   
10. # ~/.my.cnf to set user-specific options.
    
11. #
    
12. # In this file, you can use all long options that a program supports.
    
13. # If you want to know which options a program supports, run the program
    
14. # with the "--help" option.
    
15. 
    
16. # The following options will be passed to all MySQL clients
    
17. [client]
    
18. #password        = your_password
    
19. port                = 3306
    
20. socket                = /var/lib/mysql/mysql.sock
    
21. 
    
22. # Here follows entries for some specific programs
    
23. 
    
24. # The MySQL server
    
25. [mysqld]
    
26. port                = 3306
    
27. socket                = /var/lib/mysql/mysql.sock
    
28. skip-locking
    
29. key_buffer = 1024M
    
30. max_allowed_packet = 4M
    
31. table_cache = 10240
    
32. sort_buffer_size = 6M
    
33. read_buffer_size = 6M
    
34. read_rnd_buffer_size = 8M
    
35. myisam_sort_buffer_size = 128M
    
36. thread_cache_size = 128
    
37. query_cache_size= 512M
    
38. # Try number of CPU's*2 for thread_concurrency
    
39. thread_concurrency = 4
    
40. back_log = 200
    
41. max_connections = 1024
    
42. table_cache = 512
    
43. read_rnd_buffer_size = 8M
    
44. thread_cache_size = 128
    
45. wait_timeout= 20
    
46. 
    
47. # Don't listen on a TCP/IP port at all. This can be a security enhancement,
    
48. # if all processes that need to connect to mysqld run on the same host.
    
49. # All interaction with mysqld must be made via Unix sockets or named pipes.
    
50. # Note that using this option without enabling named pipes on Windows
    
51. # (via the "enable-named-pipe" option) will render mysqld useless!
    
52. #
    
53. #skip-networking
    
54. 
    
55. # Replication Master Server (default)
    
56. # binary logging is required for replication
    
57. log-bin=mysql-bin
    
58. 
    
59. # required unique id between 1 and 2^32 - 1
    
60. # defaults to 1 if master-host is not set
    
61. # but will not function as a master if omitted
    
62. server-id        = 1
    
63. 
    
64. # Replication Slave (comment out master section to use this)
    
65. #
    
66. # To configure this host as a replication slave, you can choose between
    
67. # two methods :
    
68. #
    
69. # 1) Use the CHANGE MASTER TO command (fully described in our manual) -
    
70. #    the syntax is:
    
71. #
    
72. #    CHANGE MASTER TO MASTER_HOST=<host>, MASTER_PORT=<port>,
    
73. #    MASTER_USER=<user>, MASTER_PASSWORD=<password> ;
    
74. #
    
75. #    where you replace <host>, <user>, <password> by quoted strings and
    
76. #    <port> by the master's port number (3306 by default).
    
77. #
    
78. #    Example:
    
79. #
    
80. #    CHANGE MASTER TO MASTER_HOST='125.564.12.1', MASTER_PORT=3306,
    
81. #    MASTER_USER='joe', MASTER_PASSWORD='secret';
    
82. #
    
83. # OR
    
84. #
    
85. # 2) Set the variables below. However, in case you choose this method, then
    
86. #    start replication for the first time (even unsuccessfully, for example
    
87. #    if you mistyped the password in master-password and the slave fails to
    
88. #    connect), the slave will create a master.info file, and any later
    
89. #    change in this file to the variables' values below will be ignored and
    
90. #    overridden by the content of the master.info file, unless you shutdown
    
91. #    the slave server, delete master.info and restart the slaver server.
    
92. #    For that reason, you may want to leave the lines below untouched
    
93. #    (commented) and instead use CHANGE MASTER TO (see above)
    
94. #
    
95. # required unique id between 2 and 2^32 - 1
    
96. # (and different from the master)
    
97. # defaults to 2 if master-host is set
    
98. # but will not function as a slave if omitted
    
99. #server-id       = 2
    
100. #
     
101. # The replication master for this slave - required
     
102. #master-host     =   <hostname>
     
103. #
     
104. # The username the slave will use for authentication when connecting
     
105. # to the master - required
     
106. #master-user     =   <username>
     
107. #
     
108. # The password the slave will authenticate with when connecting to
     
109. # the master - required
     
110. #master-password =   <password>
     
111. #
     
112. # The port the master is listening on.
     
113. # optional - defaults to 3306
     
114. #master-port     =  <port>
     
115. #
     
116. # binary logging - not required for slaves, but recommended
     
117. #log-bin=mysql-bin
     
118. 
     
119. # Point the following paths to different dedicated disks
     
120. #tmpdir                = /tmp/               
     
121. #log-update         = /path-to-dedicated-directory/hostname
     
122. 
     
123. # Uncomment the following if you are using BDB tables
     
124. #bdb_cache_size = 64M
     
125. #bdb_max_lock = 100000
     
126. 
     
127. # Uncomment the following if you are using InnoDB tables
     
128. #innodb_data_home_dir = /var/lib/mysql/
     
129. #innodb_data_file_path = ibdata1:10M:autoextend
     
130. #innodb_log_group_home_dir = /var/lib/mysql/
     
131. #innodb_log_arch_dir = /var/lib/mysql/
     
132. # You can set .._buffer_pool_size up to 50 - 80 %
     
133. # of RAM but beware of setting memory usage too high
     
134. #innodb_buffer_pool_size = 256M
     
135. #innodb_additional_mem_pool_size = 20M
     
136. # Set .._log_file_size to 25 % of buffer pool size
     
137. #innodb_log_file_size = 64M
     
138. #innodb_log_buffer_size = 8M
     
139. #innodb_flush_log_at_trx_commit = 1
     
140. #innodb_lock_wait_timeout = 50
     
141. 
     
142. [mysqldump]
     
143. quick
     
144. max_allowed_packet = 16M
     
145. 
     
146. [mysql]
     
147. no-auto-rehash
     
148. # Remove the next comment character if you are not familiar with SQL
     
149. #safe-updates
     
150. 
     
151. [isamchk]
     
152. key_buffer = 384M
     
153. sort_buffer_size = 6M
     
154. read_buffer = 2M
     
155. write_buffer = 2M
     
156. 
     
157. [myisamchk]
     
158. key_buffer = 384M
     
159. sort_buffer_size = 6M
     
160. read_buffer = 2M
     
161. write_buffer = 2M
     
162. 
     
163. [mysqlhotcopy]
     
164. interactive-timeout
     
165. wait_timeout= 20
     
166. interactive_tiomeout=120

复制代码

mysql配置

各位大家有什么意见没

hackroad

phpmyadmin下 show processlist 发现

  ID   用户   主机   数据库   命令   时间   状态   SQL 查询  
Kill   1828184   root   localhost   mlkz   Query   1   Sorting result   SELECT t . * , f.name
FROM cdb_threads t, cdb_forums f
WHERE f.fid = t.fid
AND f.fid NOT
IN ( 24, 160, 106 )   
Kill   1828186   root   localhost   mlkz   Sleep   0   ---   ---  
Kill   1828187   root   localhost   mlkz   Query   3   Sorting result   SELECT t . * , f.name
FROM cdb_threads t, cdb_forums f
WHERE f.fid = t.fid
AND f.fid NOT
IN ( 24, 160, 106 )   
Kill   1828190   root   localhost   mlkz   Query   1   Sending data   SELECT p . * , m.uid, m.username, m.groupid, m.adminid, m.regdate, m.lastactivity, m.posts, m.digestpos

就是在这个时候 论坛会卡住

emmoblin

是不是mysql可以设置一些优化？
感觉像首页这种应该可以缓存长时间的

bruce1000

我以前用Centos带DZ的论坛也是很慢，后来没找出问题，直接换win2003server比centos快多了

jiang313hua

优化下mysql缓存看看。
看到你的my.cnf是memory = 512M  的配置，换成large模式的cnf试试。

crastyl

死楼主没优化好